Microsoft Azure AZ-800 — Section 18: Configure and manage Windows Server file shares Part 3

141. Configure BranchCache

I’d like to now help you in understanding the concept of a feature called Branch Caching Now branch caching has been around since Windows seven came out and is supported by Windows seven and higher. And on the server side, you have to have at least a Windows Server 2008 R2 or higher.

OK, so Windows seven or higher and Windows Server 2008 R2 or higher. Those are the combinations of Windows machines that support this, and the goal of branch caching is to allow data to be cached on machines and then shared around with other machines on the network so that it can conserve bandwidth across your network.

So let me help you kind of visualize this. Let’s let’s draw some things out here.

So first off, we’ll say that we have three sites.

OK, maybe the first site is in New York, and then maybe we’ve got Dallas, Texas, and then maybe we’ve got like Orlando, Florida.

OK.

So, we’ll say New York is like our headquarters.

OK, we’ll say this headquarters HQ. And then maybe this Dallas, Texas. And then maybe this Orlando, Florida. All right. And we’ve got connections that are connecting these sites together. And maybe these connections are, you know, their way in connections.

So, they’re not they may not be ultimately, you know, super duper fast because they’re there. And, you know, maybe you were.

So, we’ve got some servers up here. All right. Various servers up here that are stored data. All right. Stored data. All right. And of course, we’ve got clients that are in these various locations here. And maybe.

Some clients and servers and things in Orlando, OK, so this kind of how we’re we’re looking at this from this perspective of some different approaches to dealing with branch caching.

So ultimately right now with the way things are whenever a user like here in Orlando is trying to get a file.

OK. The problem is that file, OK, that file is located up in New York, right? Maybe, on this little server right here. We just kind of zoom in on that will make that look like it’s a it’s a file. It’s got it’s got writing on it.

OK.

So this our writing some kind of document or something.

OK. Could be an image. Could be a document. This case is going to be a document, OK? And with the way things are right now, you got users, let’s say, in Orlando, in Dallas that occasionally need to need to connect up and they need to access that.

So this user here is going to go and access that file and then it’s going to pull the file down. And that’s great. Wonderful. This user now gets access to it, but they have, you know, might be a little bit slow because they’ve got to go across the way in. Well, then the next user does the same thing. It’s going to happen the same way. It’s going to basically have to download that file for that same user, for that other user.

So now with bridge caching, there’s two different approaches to it. The first approach is what we call posted cache mode.

OK, hosted cache mode means that you’re going to have a server. This server right here, we might call it a hosted cache server. Hosted cache server. All right. Again, it’s got to be at least Windows Server 2008 or higher 2008 or two or higher. That’s our host to her and we can implement that by using a group policy. We can basically say, OK, you’re going to be a hosted cash server, and we can also make sure that that’s installed on the server. You can install it through server manager and we can we can tell everybody about it through group policies.

So the clients will know about it.

So now what ends up happening is you have the first serve as the first user that’s going to go up and access to follow before any devices will try to access the file. It’ll always check with the hosted cache server first.

So, if not, the user is going to go up there and grab this file when the user grabs the file. What’s really nice about that is that the user is going to store a copy of it in the hosted cache server.

So the next user that goes to access to file is going to check the host of cache services. If it’s there and it is, the user can just grab it. This all transparent to the users. The users don’t even know this going on.

So, it’s a beautiful thing. The other option for dealing with branch caching is to use what is called distributed cache mode.

OK, now distributed cache mode does not require any servers to be in the site.

So, if you had a smaller site or something with just a bunch of machines, then with distributed cache mode, what’ll happen is before a computer will go across the way in, it will announce that it’s looking for that document so these machines canspecify whether or not they have it or not. The machines will keep metadata on who’s got what documents recently, and if most, the client will just check to see if the document is changed.

So this client will go up and he’ll grab this document, right? And so when he grabs a document, he’ll cache a copy of it on his machine.

So he’ll announce also that he’s got the document. The clients over here will have metadata that says he’s got the document.

So the next person that needs the document can actually get it directly from this device. They’ll also occasionally check to see before they pull the document and look to see if there’s been a change.

So that is what distributed cache mode is ultimately hosted. Cache mode is generally the one people say is the best solution, but it does require you to have a hosted cache server. But the benefit? I think the benefits are obvious here. The benefit to using branch caching is you’re going to conserve bandwidth by utilizing branch caching and basically be able to speed the performance up, especially if you got slow wind connections.

So that’s the idea of what we use branch caching for. Let’s see how we can set up branch caching on this server here in one C DC one.

So first thing we’re going to do is, let’s say we’re going to maybe do hosted cache mode on NYC DC one. I’m going to go to server manager and I’m going to go to manage ADM’s and features. We’re going to go to the roles page here, expand out file and storage services, file and I scuzzy services, and we’re going to enable branch cache and network file or for network file.

So, we’re going to click next on there. All right, next, we’re going to go ahead and install in our polls recording wall that’s being installed.

OK, so once the installs done, I’m going to hit close and at that point I’m now ready to go into group policies and point everybody to my server as a hosted cache server.

So, I’m going to go to ols and go down to group policy management. All right. And from there, we’re just going to create a group policy object here. Click new I’m just going to call it branch cash. All right, I’m going to edit that GPO and we’re going to go underneath computer config policies administrator templates network and you’ll see a folder called Branch Cache right here.

So, we’ll go to that. And to turn on branch caching for your clients, you’re going to use this option here, you’re going to turn it to enable and then to set the mode. I’m not going to if I want to use distributed caching mode, I can set this here. You can actually use both. You can actually turn on distribute cache mode or host a cache mode. And if clients don’t have a host of cache server, they’ll use distributed cache. But I’m going to do hosted cache mode here. Click Enabled. We’re going to then go here, put NYC -DC one as our server. We’re going to click OK. At that point, we don’t actually have to configure any of this other stuff here unless we want to. We can close out of the GPO drag and drop it over domain, over our domain, I should say. In my practice, .com, and we’ve now got that going at that point, we would run GP update clients, of course, would take 90 to 120 minutes to refresh, but we’ve now got branch caching set up and you can test that out. View event logs and all that and see that it’s working, but ultimately it’s pretty easy to set up and very easy to configure.

142. Visualizing Distributed File System (DFS)

It’s now time for us to discuss a concept that better in the Windows operating system or the Windows Server operating system for quite a while now. There was a original version of it that was sort of out in in even the late 90s. And then Microsoft eventually released what is officially called DFS in the end when Windows 2000 came out.

So you can say that this definitely well over a couple of decades old and now there’s been a few improvements to it. For the most part, it stayed the same system all these years.

So distributed file system is it’s a multi-tier based system that allows you to help centrally control all of your different share resources.

So, if you think about users having to make connections to all these different servers to access the things they need, they need, you know, to gain access to for their job. Generally speaking, you might have lots of servers. You may have five, 10, 15 servers, and you don’t want these users to have to keep track of what servers they need to go to for what particular resource.

So one of the things distributed file system can do is help with that. With making it where we have a central place, people go to find what they need and then be directed where they need to go. Another another thing about distributed file system is it can replicate files as well, so you can use that as a way of, you know, having your files backed up or making copies of files collecting files from a certain office, things like that.

So let’s take a deeper look at what this all about.

OK, so the first thing we’ll look at if we think about distributed file system. Is let me just draw a couple of folders here. All right. All right. All right, and then so from there, we’ll have a server. All right. And then maybe let’s say that we have let’s just say we have four servers.

OK.

OK, so ordinarily, if you had four servers and you have, you know, different shares on these different servers, you would your users would have to know about what the servers names were.

So, for example, let’s just say this one for server one, this asked to for Server two, three for Server three and four for Server four.

So from there, you might have some shares in there and each one of these that go to different departments.

OK, like I might have a folder for sales, for H.R, for accounting. Let me just kind of lower my font a little bit so, I can fit this in here. But like, you know, sales will say this sales one. This H.R. one. This, we’ll say finance one. All right.

So that’s. Not quite fitting in there, but. That’ll work. And then we’ve got sales to H.R. two. All right, finance to so on and so forth. All right.

So you kind of get the idea here and I’ll explain why I’m doing this. But you know, the issue we run into here again, though, is that your users may have these different shares in different on different servers and you’re ultimately your users having to keep up with all of these different servers. Maybe, they’ve got matt drives to him, or maybe they have to just browse over the network.

So, you know, a user who is in sales is like, Well, I got four different servers. I got to keep up with. Same for H.R, same for finance.

Now I’m not saying this. This will be the exact scenario in your office, but you could have a basically a bunch of servers that your users are needing to remember to go to in order to get their resources.

So, you know, if a user in sales needs to get to, let’s say, sales one and then they’re having to they’re having to put in the you and CPF, the universal naming convention slash slash S1 slash sales one, or they’re having a browse across the network to S1 to get there now.

So one of the great things about DFS, there’s a couple of different approaches to it. The first approach to DFS is what we call a namespace DFS.

OK, so with namespace DFS, we have a DFS server, so, we have a server that has DFS installed on it. You distributed file system. And from there, what we could actually do is we can create these parent folders. All right.

So, I can have a parent folder here for sales. And another parent folder for me. Just move all this up a little bit. Another parent folder for H.R. or another for finance.

So on and so forth. And ultimately, what these folders can do is they can act as links almost like a shortcut to get to see all of the different folders for the different departments.

So what would happen here is I can actually have inside of these parent folders I can have, it’ll create links for each one of these folders.

So, if you went into sales, you would essentially have a link that links to sales. One links to sales to links to sales, three links to sales, four so on and so forth.

So a user is only needing to remember the name of this one server and they can get to everything, all of their different folders.

So, so, you know, user could even have a mapped drive. You know, I could have a z drive that basically maps to slash slash DFS, whatever the DFS servers, names like sales or H.R. or finance. And the users can basically get into the folder and they can see everything is related to their department.

OK, now the thing is you’re not using replication by doing that, by doing what’s called the namespace method. You’re just these are basically like links, and they’re a little better than shortcuts because users can actually drag and drop into them and stuff like that, which is cool.

So, if a user had a map, drive sales or H.R. or finance, they could actually open those up. They can see these child folders in there and they can drag and drop into them, and it goes to where it needs to go. Links to where it needs to link.

So essentially what you got is you’ve got a link between your DFS servers to these other servers. That’s what the namespace method of DFS does.

Now, the other thing we can do is we can do something called DFS R DFS R is distributed file system replication.

OK, now a distributed file system replication, you can actually have the folders replicate.

OK, so the contents of sales one can go in the contents of a sales one folder that’s inside this parent. The contents of H.R. one could go into the contents of H.R. one as a parent.

So again, if you opened up this H.R. one, you actually see and you’re sorry, if you open up this whole parent folder, you would see all the H.R. folders and it could actually replicate all of those.

OK.

Now, another alternative to that of using using that is, you know, you could do that for centralization or you could also do it to have multiple copies in different offices.

So what I mean by that is like, let’s say that these ovals that I’m drawing are sites and you got maybe a site in New York City and you’ve got a site in Dallas, Texas, OK? I can have file servers in each of these offices that are replicating with each other.

So you know, you have a DFS one server here. You can have a DFS two server here and you can have a folder that’s actually replicating between each server if you wanted to. And so this great, too. In what’s awesome about this all gets stored in Active Directory. When you do this, you can you can store this all in Active Directory if you want. And when you do that, this share.

OK, like if I named the share, we’ll call it sales, OK? Sales. It can be replicated between these different locations, and what’s really neat about that is when a user accesses it, they can access it by the domain name backslash backslash in my case, examlabpractice.com slash sales. And what’s really cool is with the help of DNS and Active Directory, the user will be routed to whichever version they’re closest to.

So, if you’re dealing with a user who’s in New York, they’re actually going to obviously go to this one. If they’re in Dallas, they’re going to go to this one, right? But what’s really cool about this that it supports some failover.

So, if for some reason a win win, a client actually queries this link and talks to Active Directory and talks to DNS, they it actually tells them about both of these servers, and it puts them in the order that is closest to the client.

So what’s needed is, of course, they’re going to go to this one of their New York, but if this one goes down, the client will automatically go here. All right. And so again, also, if they make any changes to the folder, the folders can synchronize.

OK, so really cool. And this can be used for allowing users in different offices to update. The same data can also be used for data collection.

So, you know you want one servers to kind of collect data from another server, or maybe you can use it for backup. There’s different reasons for that, but that is the DSS. DFES are method of DFS, so those are the two main ways that a distributed file system are used. First namespace and then also distributed file system replication. The FSR is going to use Active Directory to assist in that replication. But all right, so hopefully that gives you now a good visualization of how DFS works.